Unlock the Editor’s Digest for free
Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
This article is an on-site version of our Inside Politics newsletter. Sign up here to get the newsletter sent straight to your inbox every weekday
Good morning. I’m pleased to be able to include something the FT’s data team have been working on for a while — our new opinion poll tracker.
The particularly cool bit, and the part I’m most excited to show you, is the way that our poll tracker depicts a “normal-sized opinion poll error”: giving us a visual depiction of the usual uncertainty about whether the polls are right or not.
But even a normal-sized opinion poll error wouldn’t change the fact the Conservative party is on course for a heavy defeat as it stands. Might further tax cuts shift the balance? Some thoughts on a striking intervention from Jeremy Hunt, while Jonathan Vincent, our in-house political data scientist, explains our poll tracker in more detail below.
Inside Politics is edited by Georgina Quach. Read the previous edition of the newsletter here. Please send gossip, thoughts and feedback to insidepolitics@ft.com
Responsible Budget?
The party’s top strategists can hardly admit that they are doing anything other than fighting to win the next election (even if the underlying mission is about minimising losses).
If one of Rishi Sunak’s aides went around telling all and sundry “let’s face it, the party is doomed”, then there would be stories in the papers about how even Downing Street thinks the game is up and Conservative MPs would panic. Self-destructive weeks like the one the Tory party has just endured would become even more frequent than they are already.
But in his column this week, Chris Giles proposes a fun and prescient way to gauge whether Sunak really thinks he can win: how he uses any extra wriggle room which may be provided in the Office for Budget Responsibility’s forecasts (these will frame the March Budget). Measures in the Autumn Statement in November left the chancellor with £13bn of “headroom” against his crucial fiscal rule, which requires the ratio of public debt to GDP to be falling in five years’ time.
If Sunak saves none of his coming war chest and blows the lot on tax cuts, the survival statistic will be zero, demonstrating that he knows his party’s chances at the election are minimal. That would be the actions of someone who merely wants to make life difficult for Labour. On the other hand, if he saves most of any fiscal windfall, the higher statistic would demonstrate that he believes he might be in power and able to effect change up to 2029.
Semi-relatedly, Jeremy Hunt has signalled that he will cut taxes further in his March Budget. Anyway, this seems like a good moment to show you our opinion poll tracker:
I think Chris is bang on the money in saying that the government’s willingness to cut taxes is a good indicator of what it actually thinks is going to happen at the next election. And I also agree with Chris in thinking these tax cuts will make the next government’s life more difficult.
There are a number of overlapping crises in the public realm — this week, Peter Foster revealed one in what had been a comparative advantage for the UK, the universities sector — and a number of geopolitical storms, all of which are going to involve spending more money. Jeremy Hunt’s possible tax reductions today would have to be clawed back by the chancellor after the next election, whether that chancellor is Hunt, Claire Coutinho or Rachel Reeves. That’s even before you get on to the expensive pet projects of the main party leaders, such as Rishi Sunak’s laudable but costly ambition for all pupils in England to study maths to 18.
But I don’t think that these tax cuts are quite about “merely making Labour’s life difficult”: the opinion polls are really, really, really very bad for the Conservative party at the moment. What big tax cuts will tell us is that the apocalyptic results suggested by our poll tracker concur with the Conservative party’s own data. Tory self-interest will be a big part of further tax cuts: but the motive I think is much more about survival than spite.
The Uncertain Kingdom
Here’s Jonathan Vincent, the FT’s political data scientist, on the poll tracker team’s methodology: The FT has launched our UK poll-of-polls, which will be continuously updated until the general election. Our tracker includes the headline voting intention figures, as well as rolling breakdowns of party support by region, age, gender, social class and previous voting behaviour.
When designing our polling average, we wanted it to be sensitive to real changes in public opinion, unaffected by random noise in the data, and not overly affected by a single pollster.
To ensure that our average is sensitive to recent events, we weight newer polls more heavily than older ones. So a poll released today is fully weighted, while one older than 30 days is discarded entirely.
To counteract random noise, we also weight each poll by the square root of its sample size, so that a poll with 4,000 respondents contributes twice as much to our average as a poll with 1,000 respondents. This reflects the way the accuracy of a poll increases with sample size.
Finally, to avoid giving undue influence to a single pollster, we only take the most recent poll from each polling company.
Even with these adjustments, there is always a chance that the entire polling industry is systematically underestimating or overestimating support for a given party, as it did in 1992 and 2015. To reflect this, our poll tracker includes a shaded region around each party’s polling average that reflects the range of likely vote shares each party would win if the election was held tomorrow.
We have chosen the width of this region by analysing polling and election results in both the UK and the US since the 1980s. Our uncertainty interval is not a traditional confidence interval, but it is a reasonable estimate of the likely range of outcomes, assuming typical levels of polling industry error. Send us an email at elections@ft.com to chat more!
Now try this
(Georgina) Here’s a web game that feels very FT: Tradle — guess the country based on its exports. It is now one of my daily habits, alongside the original “Wordle”, “Worldle” and the NYT’s “Connections”. See also the FT quiz on the week’s current affairs.
Have a lovely weekend, however you spend it.
Top stories today
Recommended newsletters for you
One Must-Read — Remarkable journalism you won’t want to miss. Sign up here
FT Opinion — Insights and judgments from top commentators. Sign up here